🔭 astrophysics

Searching for signatures of self-interacting dark matter in halos from full-physics simulations: From 3D structure to projected observables

This study utilizes full-physics simulations to demonstrate that while baryonic and projection effects generally suppress self-interacting dark matter (SIDM) signatures in galaxy groups, weak lensing and kinematic analyses of massive clusters offer promising avenues for detecting SIDM imprints, potentially revealing up to 20% deviations in shear profiles for halos exceeding $10^{13}\ \text{M}_\odot$ .

Original authors: Giovanni Y. Ferron, Antonio Ragagnin, Lorenzo Pizzuti, Moritz S. Fischer

Published 2026-08-25

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Giovanni Y. Ferron, Antonio Ragagnin, Lorenzo Pizzuti, Moritz S. Fischer

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

For decades, the prevailing story of the universe has relied on a mysterious substance called dark matter. This invisible material does not emit light, yet its gravity holds galaxies together, acting as the cosmic scaffolding that keeps the visible universe from flying apart. In the standard version of this story, dark matter is "cold" and "collisionless," meaning its particles pass right through one another like ghosts, interacting only via gravity. This idea has been incredibly successful at explaining the large-scale structure of the cosmos. However, when astronomers look closely at the centers of smaller galaxies, the standard model sometimes struggles to match what they see. To fix these small-scale mismatches, scientists have proposed an alternative: self-interacting dark matter. In this version, dark matter particles can bump into each other and exchange energy, much like gas molecules in a room, which would change how they clump together in the hearts of galaxies.

The question of whether dark matter behaves like a ghost or a gas has moved from theory to the realm of observation, thanks to powerful new telescopes and supercomputers. Researchers are now trying to find the subtle fingerprints of these collisions in the real universe. The challenge is that the effects are often tiny and easily hidden by other cosmic processes. A team of astronomers has recently tackled this problem by running massive, detailed simulations of galaxy groups and clusters. They wanted to see if the signatures of self-interacting dark matter could survive the messy reality of the cosmos, where normal matter like gas and stars also plays a major role, and whether future surveys could actually spot these differences.

The researchers began by building a digital universe using a supercomputer. They created simulations of galaxy clusters, which are the largest structures in the universe held together by gravity, containing hundreds or thousands of galaxies. They ran these simulations twice: once with the standard, ghost-like dark matter, and once with the self-interacting version where particles could collide. Crucially, they did not just simulate the dark matter; they included the normal matter as well, modeling how gas cools, forms stars, and interacts with the dark matter. This "full-physics" approach is essential because the presence of stars and gas can drastically alter the shape of a galaxy cluster's center, potentially masking the effects of dark matter collisions. They focused on two specific types of self-interacting models: one where collisions happen rarely but with significant force, and another where collisions happen very frequently but with very gentle nudges.

When the team analyzed the results, they found that the presence of normal matter made the job of detecting self-interacting dark matter much harder than expected. In the simulations where only dark matter existed, the self-interacting version created a distinct "core" in the center of the galaxy clusters, making the density drop off more gently than in the standard model. However, once they added the gas and stars, the picture changed. The normal matter pulled the dark matter inward, making the center denser and sharper. This effect, known as adiabatic contraction, largely canceled out the softening caused by the dark matter collisions. As a result, for most galaxy clusters in their simulations, the difference between the standard model and the self-interacting model became very small, often less than five percent. This suggests that simply looking at the total mass distribution of a cluster might not be enough to tell the two theories apart.

The researchers then looked at the problem from a different angle, simulating what an observer on Earth would actually see. Because we view these clusters from a single direction, we see a flattened, two-dimensional projection of their three-dimensional structure. This projection tends to wash out the subtle differences even further. However, the team discovered a promising exception. When they focused on the very most massive clusters in their simulations, those with masses exceeding ten to the power of fifteen times the mass of our sun, the self-interacting models behaved differently. Instead of creating a soft core, these massive clusters developed a center that was even denser and sharper than the standard model. This counter-intuitive result, driven by the complex interplay between the dark matter collisions and the heavy concentration of normal matter, offered a new potential signature.

To test if this signature could be seen in the real world, the team compared their most massive simulated clusters to actual observations of two famous galaxy clusters, MACS J1206 and Abell S1063. These real clusters have been studied extensively, with astronomers measuring the speeds of the galaxies inside them to map out their mass. The researchers found that the uncertainties in these real-world measurements are now small enough that they could potentially distinguish between the standard model and the self-interacting models. If the self-interacting theory is correct, the cores of these massive clusters should look slightly different than what the standard model predicts, and current data might already be precise enough to catch that difference.

Finally, the team investigated a different method of detection: weak gravitational lensing. This phenomenon occurs when the gravity of a massive cluster bends the light from galaxies behind it, distorting their shapes. By measuring these distortions, astronomers can map the mass of the cluster without needing to see the galaxies inside it. The simulations showed that for clusters with masses around ten to the power of thirteen times the mass of the sun, the weak lensing signal could reveal deviations of up to twenty percent between the standard and self-interacting models. This is a significant difference, suggesting that future surveys, which will map thousands of clusters with extreme precision, could use this technique to test the nature of dark matter.

The study concludes that while the search for self-interacting dark matter is complicated by the presence of normal matter and the way we view the universe, it is not impossible. The effects are not erased, but they are shifted to specific scales and specific types of massive structures. For smaller clusters, the signal is faint and easily hidden, but for the most massive clusters and through the lensing of background light, the fingerprints of self-interacting dark matter remain visible. The work provides a roadmap for future observations, telling astronomers exactly where to look and what to measure. It suggests that by combining data on the motion of galaxies with the subtle distortions of light, we may soon be able to confirm whether dark matter is a silent ghost or a substance that bumps and interacts with itself, fundamentally changing our understanding of the universe's invisible backbone.

Technical Summary: Searching for signatures of self-interacting dark matter in halos from full-physics simulations

Problem Statement
The nature of dark matter (DM) remains an open issue in cosmology. While the standard $\Lambda$ CDM paradigm assumes DM is collisionless, Self-Interacting Dark Matter (SIDM) models propose that DM particles interact via non-gravitational forces, potentially resolving small-scale structure issues like the "core-cusp" and "missing satellites" problems. However, previous theoretical studies have largely focused on dark-matter-only (DMO) simulations or semi-analytic approaches, often neglecting the complex interplay between baryonic physics (gas and stars) and DM self-interactions. Furthermore, there is a need to determine which projected observables (2D) best retain the signatures of SIDM, as line-of-sight projection and baryonic effects may wash out the 3D structural differences predicted by theory. This work addresses the gap by investigating SIDM imprints on galaxy groups and clusters using full-physics hydrodynamical simulations, specifically evaluating the viability of weak lensing and kinematic mass measurements as probes.

Methodology
The authors utilize a suite of six "zoom-in" hydrodynamical simulations from the DIANOGA project, covering galaxy clusters with virial masses $M_{200} \in [10^{14}, 10^{15}] M_\odot$ . Each region is simulated under three DM scenarios: standard Cold Dark Matter (CDM), rare SIDM (rSIDM), and frequent SIDM (fSIDM). Both DMO and full-physics (FP, including gas, stars, cooling, star formation, and AGN feedback) runs are performed.

The SIDM models implemented follow a velocity-dependent cross-section framework where the interaction arises from a light scalar mediator, described by a radial Yukawa potential. The models are defined by two regimes:

rSIDM: Isotropic scattering with large scattering angles.
fSIDM: Highly anisotropic scattering with small deflection angles.
Both are normalized to a momentum-transfer cross-section of $\sigma_{T,0}/m_{DM} = 40 \text{ cm}^2 \text{ g}^{-1}$ at low velocities, with a transition velocity $v_c = 200 \text{ km s}^{-1}$ .

The analysis focuses on a single snapshot at redshift $z=0.4$ . The authors perform a stacking analysis of mass and density profiles for halos with $M_{200} > 5 \times 10^{12} M_\odot$ to mitigate numerical noise, while also analyzing individual profiles for the most massive halos ( $M_{200} \gtrsim 10^{14} M_\odot$ ). They fit cumulative mass profiles to the Navarro–Frenk–White (NFW) model to derive concentration and scale radii. Crucially, they compute projected quantities (2D mass, density, and surface density excess $\Delta\Sigma$ ) along three Cartesian axes to simulate weak lensing observables. Finally, the most massive simulated halos are compared directly to kinematic and lensing reconstructions of the real clusters MACS J1206 and Abell S1063.

Key Contributions and Results

Alleviation of SIDM Signatures by Baryons and Projection:
The study finds that both baryonic effects and line-of-sight projection significantly reduce the detectability of SIDM in mass and density profiles.
- 3D Profiles: In DMO simulations, SIDM produces a characteristic drop in central density (up to 10–20% below CDM) and a mass bump at intermediate radii ( $r \sim 0.3-0.4 r_{500c}$ ) due to heat transfer. However, in full-physics (FP) simulations, baryonic cooling and adiabatic contraction pull DM particles into the core, steepening the inner profile. This "alleviates" the SIDM core-forming behavior, reducing deviations from CDM to $\lesssim 5\%$ in lower-mass halos.
- Projected Profiles: Projecting the 3D profiles further washes out SIDM signatures. In the FP case, projected mass and density profiles show deviations of only $\lesssim 10\%$ at best, becoming indistinguishable from CDM at intermediate radii.
Weak Lensing as a Promising Probe:
Despite the suppression in mass profiles, the authors identify the surface density excess ( $\Delta\Sigma$ ), the primary observable for weak lensing, as a more sensitive probe.
- For halos with $M_{200} \gtrsim 10^{13} M_\odot$ , the $\Delta\Sigma$ profiles exhibit deviations of up to $\sim 20\%$ (above the resolution limit) in the FP case.
- This suggests that while standard mass reconstruction might miss SIDM, weak lensing shear profiles could potentially detect these deviations in cluster-sized halos, provided systematics are controlled.
Behavior of Massive Halos and Comparison to Observations:
Contrary to the core-formation trend in lower-mass halos, the most massive simulated halos ( $M_{200} \sim 10^{15} M_\odot$ ) in the FP runs exhibit cuspier inner profiles than CDM.
- The paper notes that this behavior is unlike that observed in lower-mass halos and attributes the cuspier profiles specifically to the presence of baryons in the most massive systems, where baryonic dominance can erase the positive velocity dispersion gradient required for core formation, leading to immediate collapse.
- When comparing these specific massive simulated halos to the observed clusters MACS J1206 and Abell S1063, the authors note that while individual profiles vary due to assembly history, the observational uncertainties in the cluster cores are small enough that the distinct SIDM enhancements (or suppressions) could theoretically be detected.
- The authors caution that the exact origin of the cuspier massive SIDM cores in their simulations is complex, potentially involving the interplay of baryonic dominance and velocity dispersion gradients, and may not be a simple numerical artifact.
Concentration-Mass Relation:
The study finds that the projected concentration–mass ( $c-M$ ) relation is only slightly modified by SIDM models (deviations $\lesssim 5\%$ ). The authors argue that the $c-M$ relation is likely not a robust discriminator for SIDM compared to direct analysis of surface density excess profiles, partly because the simple NFW model used to fit concentrations cannot fully capture the cored internal structures of SIDM halos, leading to biased estimates.

Significance and Claims
The paper claims to provide a necessary baseline for future SIDM studies by quantifying how baryonic physics and projection effects alter theoretical predictions. The primary significance lies in identifying specific observational windows where SIDM remains detectable despite these mitigating factors:

Weak Lensing: The authors claim that weak lensing shear profiles in clusters ( $M_{200} \gtrsim 10^{13} M_\odot$ ) offer a promising avenue for detecting SIDM, with potential deviations up to 20%.
Kinematics in Massive Clusters: They suggest that the small observational uncertainties in the cores of very massive clusters (like MACS J1206) could allow for the detection of SIDM imprints, even if the standard core-formation signature is reversed to a central density enhancement in these specific regimes.

The authors remain modest regarding their conclusions, noting that their sample of massive halos is too small for a statistically robust stacking analysis and that future work with higher-resolution simulations and larger samples is required to confirm the behavior of the most massive clusters. They also highlight that current results rely on idealized scenarios where observational systematics are neglected, and future modeling of these systematics is essential for realistic detection prospects.

Technical Summary: Searching for signatures of self-interacting dark matter in halos from full-physics simulations

More like this